Abstract
Currently, performance analysis on complex manufacturing systems is performed in an ad hoc way since the main objective is to verify if the strategy designed has been helping companies achieve their targets following a reactive approach. However, more and more companies are performing in competitive markets, forcing them to become more proactive than reactive. This way, a simple approach is no longer suitable for this type of companies, and a stronger and effective interaction between the strategic and operational layers is key. Therefore, this research proposes a framework composed of both qualitative and quantitative methods that allow decision-makers to better understand their production system. Moreover, using key leading indicators as reference, the idea is to provide companies with the ability to anticipate future performance behaviors based not only on the knowledge acquired, but also on a mathematical tool that will synthesize this knowledge and infer future performance behaviors. This paper explores a critical issue for contemporary industrial organizations and sustainability issues concerning energy consumption. In this scope, an important research was performed aiming at modeling and understanding the normal behavior of electricity consumption, as well as the factors affecting energy consumption in the painting line of an automotive plant. © Springer International Publishing Switzerland 2013.

Abstract
This paper proposes a new model to predict company failure in the construction industry. The model includes three major innovative aspects. The use of strategic variables reflecting the key specificities of construction companies, which are critical to explain company failure. The use of data mining techniques, i.e. support vector machine to predict company failure. The use of two different sampling methods (random undersampling and random oversampling with replacement) to balance class distributions. The model proposed was empirically tested using all Portuguese contractors that operated in 2009. It is concluded that support vector machine, with random oversampling and including strategic variables, is a very robust tool to predict company failure in the context of the construction industry. In particular, this model outperforms the results obtained with logistic regression.

Abstract
In this paper we present three hybrid heuristics for the single machine scheduling problem with quadratic earliness and tardiness costs, and no machine idle time. Our heuristic is a combination of a steady-state genetic algorithm and three improvement procedures. The two computationally less expensive of these three improvement procedures are used inside the genetic algorithm to improve the schedule obtained after the application of genetic operators, whereas the more expensive one is used to improve the best solution returned by the genetic algorithm. We have compared our hybrid approaches against existing recovering beam search and genetic algorithms. The computational results show the effectiveness of our hybrid approaches. Indeed, our hybrid approaches outperformed the existing heuristics in terms of solution quality as well as running time.

Abstract
This paper applies network analysis to study the evolution of aviation networks. It takes a different approach from previous research that usually only explores airline networks. The aviation network of the airports of Lisbon, Faro and Porto, is modelled using the supply of seats and the passenger demand between 2001 and 2010. This analysis is complemented with a study of the commercial evolution of the three airports. It is noticeable the impact of low-cost carriers in the evolution of the configuration of the network over time. A de-concentration effect is also shown to occur, due to the interaction between airport and airline decisions that favour network development. The results of the analysis highlight how critical it has become for airport managers to assess and satisfy the real requirements of the different types of airlines, in order to reduce uncertainty and increase traffic.

Abstract
In this paper, we propose a multiobjective local search metaheuristic for a mean-risk multistage capacity investment problem with process flexibility, irreversibility, lumpiness and economies of scale in capacity costs. In each period, discrete decisions concerning the investment in capacity expansion, and continuous decisions concerning the utilization of the available capacity to satisfy demand are considered. We solve the capacity utilization problems with linear programming, in order to find the minimum capacity for each resource with the other resources remaining unchanged, this way providing information on the feasibility of the discrete investment decisions. Conditional value-at-risk is considered as a risk measure. Results of a computational study are presented, that show the approach is capable of obtaining high-quality approximations to the efficient sets, with a modest computational effort.

Abstract
Regular and non-regular production can often be found in multipurpose batch plants, requiring two distinct operating strategies: campaign and short-term production. This paper describes a sequential approach for the simultaneous scheduling of campaign and short-term products in multipurpose batch plants. Campaign products follow a periodic scheduling and are constrained to a monthly demand and a safety stock level. Shortterm products have a non-periodic scheduling and must respect tight delivery time windows. Our integrated model is based on the Resource-Task Network (RTN) representation proposed by Pantelides (1994) and uses a discrete-time formulation.